Waste Management xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Waste Management
journal homepage: www.elsevier.com/locate/wasman

Life cycle assessment of electronic waste treatment

Jinglan Hong a,b,⇑, Wenxiao Shi a, Yutao Wang c, Wei Chen a, Xiangzhi Li d,⇑
a
Shandong Provincial Key Laboratory of Water Pollution Control and Resource Reuse, School of Environmental Science and Engineering,
Shandong University, Jinan 250100, PR China
b
Shandong University Climate Change and Health Center, Public Health School, Shandong University, Jinan 250012, PR China
c
School of Life Science, Shandong University, Shanda South Road 27, Jinan 250100, PR China
d
School of Medicine, Shandong University, Jinan 250012, PR China

a r t i c l e i n f o a b s t r a c t

Article history: Life cycle assessment was conducted to estimate the environmental impact of electronic waste (e-waste)
Received 28 September 2014 treatment. E-waste recycling with an end-life disposal scenario is environmentally beneficial because of
Accepted 29 December 2014 the low environmental burden generated from human toxicity, terrestrial ecotoxicity, freshwater ecotox-
Available online xxxx
icity, and marine ecotoxicity categories. Landfill and incineration technologies have a lower and higher
environmental burden than the e-waste recycling with an end-life disposal scenario, respectively. The
Keywords: key factors in reducing the overall environmental impact of e-waste recycling are optimizing energy con-
E-waste
sumption efficiency, reducing wastewater and solid waste effluent, increasing proper e-waste treatment
End-life disposal
Improper treatment
amount, avoiding e-waste disposal to landfill and incineration sites, and clearly defining the duties of all
Recycling stakeholders (e.g., manufacturers, retailers, recycling companies, and consumers).
Heavy metal Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction environmental impacts generated by e-waste recycling are com-

Electronic waste (e-waste) refers to waste generated from dis-
carded electrical or electronic devices (e.g., cell phones, computers,
TV, printers). Given the vast technological advancement and eco-
nomic development in many countries in recent years, the volume
of e-waste produced has significantly increased (Qu et al., 2013;
Robinson, 2009). The current global production of e-waste is
around 25 million tons per year (Robinson, 2009), with the greatest
amount of e-waste imported in China (Chi et al., 2014). However,
compared with e-waste recycling in developed countries, that in
China suffers from a high occurrence of environmental pollution
and low energy efficiency. One of the most important mineral
resources, e-waste is traditionally recovered in China by workers
with the use of open flames or hot plates as a convenient way to
remove electronic components (Allsopp et al., 2006). The improper
handling of e-waste releases heavy metals (e.g., lead, cadmium,
mercury, and beryllium) and hazardous chemicals (e.g., dioxins,
furans, polychlorinated biphenyl) that seriously deteriorate the
atmosphere, water, and soil quality (Li et al., 2014; Xu et al.,
2014) and thus affect human health (Liu et al., 2009). The potential
⇑ Corresponding authors. School of Environmental Science and Engineering,
Shandong University, Jinan 250100, PR China (J. Hong); School of Medicine,
Shandong University, Jinan 250012, PR China (X. Li). Tel.: +86 (0531)88362328, +86
(0531)88382021; fax: +86 (0531)88364513.
E-mail addresses: hongjing@sdu.edu.cn (J. Hong), xiangzhi@sdu.edu.cn (X. Li).
plex and involve multi-factorial participation (e.g., process, activ-
ity, and substances). In this regard, a systematic consideration of
emission inventories and the environmental potential impacts
caused by e-waste recycling is highly needed.
Life cycle assessment (LCA) is a systematic approach to assess
and quantify the environmental performance associated with all
stages of a product creation, processes, and activities (ISO 14040,
2006). LCA can simultaneously, systematically, and effectively
evaluate and identify environmental inventory, impact, key factors,
decisions, optimization, and improvement opportunities associ-
ated with all stages of system boundary. Several studies have ana-
lyzed the environmental impact of e-waste treatment on the
environment via LCA (Song et al., 2012; Niu et al., 2012). Song
et al. (2012) investigated e-waste treatment by using emergy anal-
ysis combined with the LCA method for a trial project in Macau.
Their results showed that recovery of metals, glass, and plastic
from e-waste can generate environmental benefits. Niu et al.
(2012) compared three cathode ray tube (CRT) display treatment
scenarios (i.e., incineration, manually dismantling, and mechani-
cally dismantling) via LCA by using literature review. Their results
showed that the incineration of CRT displays has the greatest
impact, followed by mechanical dismantling. Despite their scien-
tific contributions, the aforementioned studies are unclear as to
whether direct air, water, and soil emissions from the industry site
of e-waste recycling are included in the calculation of results.
Inventory databases are also variable in terms of regionalization,

http://dx.doi.org/10.1016/j.wasman.2014.12.022
0956-053X/Ó 2015 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Hong, J., et al. Life cycle assessment of electronic waste treatment. Waste Management (2015), http://dx.doi.org/10.1016/
j.wasman.2014.12.022
2 J. Hong et al. / Waste Management xxx (2015) xxx–xxx
geography, and uncertainties involved. However, in the aforemen-
tioned studies, no regionalized database was selected to determine
the environmental effects of e-waste in China. Most data were col-
lected from European database (Ecoinvent centre, 2010). Therefore,
accurate results for Chinese case studies are difficult to obtain. The
quantification and communication of uncertainties related to LCA
results are also vital for their correct interpretation and use.
However, most LCA experts, including the authors of the aforemen-
tioned studies, still conduct LCA without considering uncertainties.
The environmental impact generated from informal recycling pro-
cesses should also be quantified because substantial e-waste in
China is recycled by individual workshops (Lin and Liu, 2012). In
this regard, the current study aims to address the aforementioned
needs, identify the key factors to improve the processes in the
Chinese e-waste recycling industry, characterize and compare
two e-waste recycling technologies commonly applied in China,
and introduce a Chinese e-waste recycling database.
2. Scope definition
2.1. Functional unit
In this study, the management of 1 ton of e-waste (i.e., com-
puter and television) is selected as the functional unit to provide
a quantified reference for all other related inputs and outputs. All
air, water, and soil emissions, raw materials and energy consump-
tion, and waste disposal are based to this functional unit.
2.2. System boundary
System boundaries were set by application of a gate-to-gate
approach. Two scenarios commonly used in China were considered
in this study, namely, e-waste treatment with end-life disposal
(ET-D) and e-waste treatment without end-life disposal (ET-ND).
Fig. 1a presents the system boundary and mass flow for the ET-D
scenario. The ET-ND scenario is simpler than the ET-D scenario
because the pollutant control system is commonly excluded in
the ET-ND scenario in many individual workshops (Fig. 1b). The
ET-D scenario involves raw materials and energy production; road
transportation of raw materials to the e-waste treatment site;
direct air, water, and soil emissions during e-waste treatment pro-
cesses (i.e., classification, disassembly, crush, electrodialysis, and
metal refining); and waste disposal (i.e., on-site wastewater and
air pollution treatment, landfill and leachates treatment, incinera-
tion). To simplify the LCA analysis of the ET-D and ET-ND scenarios,
the common process of e-waste collection to the e-waste treat-
ment site is excluded. The infrastructure (i.e., construction and
equipment) process is also excluded because of the lack of infor-
mation from selected e-waste treatment sites. Moreover, infra-
structure provides a minimal overall contribution to the potential
environmental impact (Ecoinvent centre, 2010).
2.3. Life cycle impact assessment methodology
Life cycle impact assessment (LCIA) results were calculated at
midpoint level by using the ReCiPe method (Goedkoop et al.,
2009) because the fate exposure of this model is consistent with
multimedia modeling. This method is also the most recent indica-
tor approach available in LCA analysis. It considers a broad set of 18
midpoint impact categories (i.e., human toxicity, photochemical
oxidant formation, particulate matter formation, ionising radiation,
climate change, ozone depletion, terrestrial acidification, freshwa-
ter eutrophication, terrestrial ecotoxicity, freshwater ecotoxicity,
marine eutrophication, marine ecotoxicity, urban land occupation,
natural land transformation, agricultural land occupation, water
Please cite this article in press as: Hong, J., et al. Life cycle assessment of electronic waste treatment. Waste Management (2015), http://dx.doi.org/10.1016/
j.wasman.2014.12.022
depletion, metal depletion, and fossil depletion). Normalization,
which is determined by the ratio of the impact per unit of emission
divided by the per capita world impact for the year 2000 (Wegener
Sleeswijk et al., 2008), was applied in this study to compare mid-
point impacts and analyze the respective share of each midpoint
impact to the overall impact. The complete characterization factors
and detailed methodology for ReCiPe are available on the website
of Institute of Environmental Science in Leiden University of
Nederland (http://www.cml.leiden.edu/research/industrialecolo-
gy/researchprojects/finished/recipe.html).
To determine the level of confidence in the assertion that ET-D
is more environmentally friendly than ET-ND, uncertainty analysis
is performed via Monte-Carlo analysis by using Simapro 8.0. The
geometric variation coefficient (GSD2) defined the 2.5th and
97.5th percentiles, namely, the 95% confidence interval of a proba-
bility distribution near the median l. For each unit process, the
GSD2 for all LCI parameters is defined by Eq. (1) (Ecoinvent
centre, 2010).
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2 2 2 2 2 2 2
GSD2 ¼ exp ½lnðU1 Þ þ½lnðU2 Þ þ½lnðU3 Þ þ½lnðU4 Þ þ½lnðU5 Þ þ½lnðU6 Þ þ½lnðUb Þ ð1Þ
where Ub is the basic uncertainty factor, whereas U1, U2, U3, U4, U5,
and U6, are the uncertainty factor for reliability, completeness, tem-
poral correlation, geographic correlation, other technological corre-
lation, and sample size, respectively The detailed methodology for
Monte-Carlo analysis using Simapro software is available in the
Ecoinvent report (Ecoinvent centre, 2010). Additionally, the contri-
bution of individual parameters in the life cycle of both scenarios is
identified by Eq. (2) (Hong et al., 2010a).
2 2 2 1=2
GSD2O ¼ exp ½S2I1 ðln GSD2I1 Þ þ S2I2 ðln GSD2I2 Þ þ    S2In ðln GSD2In Þ 
ð2Þ
where GSD2O, Si, and GSDi2 are the overall coefficient of variation in
the final result, the model sensitivity to each input parameter (i),
and its coefficient of variation of individual inputs, respectively.
2.4. Data sources
Operation data (i.e., energy, chemicals, raw material, water,
wastewater, solid waste, and product) and direct water and air
emissions (i.e., before and after pollutant treatment) from an
e-waste recycling site in Tianjin, China were collected to generate
a life cycle inventory for e-waste treatment (Table 1). The annual
capacity for e-waste treatment in this site, which is a professional
dismantling enterprise in northern China, is around 24 kt in 2012.
For the ET-ND scenario, the company monitoring data of the
Tianjin e-waste recycling site related to the direct air and water
emissions from e-waste classification, disassembly, crushing, elec-
trodialysis, and metal refining process before pollutant treatment
were used to generate water and air emissions. Furthermore, data
from five Guiyu e-waste dumpsite samples were aggregated to
generate the average direct soil emissions for the ET-ND scenario
(Brigden et al., 2005). Guiyu is a town located in Guangdong, China
and is one of the largest e-waste sites in the world. This town has
been extensively working in the e-waste processing business by
using primitive and hazardous methods (Sthiannopkao and
Wong, 2013; Brigden et al., 2005). It therefore represents a typical
situation for the ET-ND scenario. In addition, 2009 onsite data-
based life cycle inventory (LCI) on coal-based electricity generation
(Cui et al., 2012), theoretical LCI calculation of road transport data
(Chen et al., 2014), and 2007 onsite data-based LCI on solid waste
landfill and incineration (Hong et al., 2010b) in China were used in
this study. Relevant background data from Europe (Ecoinvent
centre, 2010), including those on chemical production, were also
collected because of the limited information on sites.
 J. Hong et al. / Waste Management xxx (2015) xxx–xxx 3

(a)

(b)

Fig. 1. System boundary and mass flow (a) ET-D scenario; (b) ET-ND scenario.

3. Results 3.2. Normalized LCIA results

3.1. LCIA results
Table 2 presents the LCIA midpoint assessment results with the
use of the ReCiPe method. The ET-ND scenario has a high potential
impact on human health, photochemical oxidant formation, terres-
trial ecotoxicity, freshwater ecotoxicity, marine eutrophication,
and marine ecotoxicity. Coal-based electricity generation signifi-
cantly contributed to the overall environmental burden for both
scenarios. Landfill and wastewater disposal showed an additional
dominant contribution to the overall environmental burden for
the ET-D scenario, whereas the additional dominant process for
the ET-ND scenario showed direct pollutant emissions as a result
of e-waste on-site disposal.
Fig. 2 shows the normalized midpoint results in each scenario.
For the ET-D scenario, the impact of climate change, human toxic-
ity, photochemical oxidant formation, particulate matter forma-
tion, terrestrial acidification, freshwater eutrophication, marine
eutrophication, freshwater ecotoxicity, marine ecotoxicity, and
fossil depletion has a dominant contribution to the overall environ-
mental impact. For the ET-ND scenario, the overall environmental
impact is mainly attributed to human toxicity, terrestrial ecotoxic-
ity, freshwater ecotoxicity, and marine ecotoxicity. For both
scenarios, the impact from the other categories has a relatively
small role. The overall environmental impact of the ET-D scenario
is relatively smaller than that of the ET-ND because of the signifi-
cant high soil emissions of the latter.

Please cite this article in press as: Hong, J., et al. Life cycle assessment of electronic waste treatment. Waste Management (2015), http://dx.doi.org/10.1016/
j.wasman.2014.12.022
4 J. Hong et al. / Waste Management xxx (2015) xxx–xxx

Table 1
Life cycle inventory of electric waste disassembling processes. Values are presented per functional unit.

Unit ET-D ET-ND

Amount Amount
Operation stage Electricity kW h 80 80
Land occupation m2 0.8 0.8
Building m2 0.4 0.4
Water g 1.03  106 1.03  106
Wastewater g 1.59  105 1.59  105
Solid waste g 1.85  105 1.85  105
Direct air emissions Particulates g 2.38 5.29  104
Nitrogen oxides g 3.48 96.99
Ammonia g 8.66  102 2.04
Hydrogen chloride g 1.53 39.24
Sulfuric acid mist g 9.75  102 2.04
Direct emissions from wastewater Nickel g 7.98  103 1.11  102
Petroleum g 1.06  102 3.31  102
Zinc g 3.19  103 2.26  102
Cyanogen g 6.4  104 6.4  104
Suspended solids g 6.39  102 3.19
Chemical oxygen demand (COD) g 1.72 5.43
Chromium g 4.79  103 4.79  103
Copper g 7.98  103 19.2
Cadmium g 7.98  103 7.98  103
Lead g 3.19  102 3.19  102
Ammonia–nitrogen g 2.63  102 5.28  102
Solid waste Antimony g 85.17
Arsenic g 4.26
Barium g 207.14
Beryllium g 2.0  104
Bismuth g 0.02
Cadmium g 7.59  103
Chromium g 29.83
Cobalt g 3.08
Copper g 1.34  103
Gold g 2.43
Lead g 318.21
Manganese g 117.23
Mercury g 0.17
Molybdenum g 3.23
Nickel g 64.98
Silver g 10.53
Tin g 67.26
Vanadium g 2.62
Yttrium g 1.25
Zinc g 512.87
Chlorinated benzenes g 1.10
Polychlorinated biphenyl g 76
Polybrominated Biphenyl Ethers g 0.42
Phthalate esters g 0.27
Aliphatic hydrocarbons g 2.85
Aromatic hydrocarbons g 2.93
Organosilicon compounds g 0.04
Organophosphate compounds g 0.04

3.3. Main contributors
The most significant contributions to LCA are shown in Fig. 3 to
further elucidate the dominant substances in the e-waste treat-
ment scenarios. For the ET-D scenario, the most dominant sub-
stances that contribute to climate change are carbon dioxide and
methane. The emissions of arsenic and selenium to water and mer-
cury to air play important roles in human toxicity. The dominant
substance in photochemical oxidant formation and marine eutro-
phication is nitrogen oxide. The emissions of non-methane volatile
organic compounds to air and nitrate and ammonium to water are
additional dominant substances in photochemical oxidant forma-
tion and marine eutrophication, respectively. In particulate matter
formation and terrestrial acidification, nitrogen oxides and sulfur
dioxide emitted to the air are the most significant substances.
The emission of particulates to the air also plays an important role
in particulate matter formation. Phosphate in water has a domi-
nant contribution to freshwater eutrophication. Vanadium, nickel,
beryllium, and selenium are the most significant substances in
freshwater and marine ecotoxicity. The emission of bromine to
water has an additional dominant contribution to freshwater
ecotoxicity. The use of coal, natural gas, and oil also significantly
contributes to fossil depletion. For the ET-ND scenario, the
substances that contributed the most to each dominant category
are direct silver and zinc to soil from the solid waste open dumping
stage. Direct copper from the waste open dumping process plays
an important role in most categories, except for human toxicity.
Direct barium and antimony emissions from the same process
are additional dominant substances in most categories, except for
terrestrial ecotoxicity.

Please cite this article in press as: Hong, J., et al. Life cycle assessment of electronic waste treatment. Waste Management (2015), http://dx.doi.org/10.1016/
j.wasman.2014.12.022
 J. Hong et al. / Waste Management xxx (2015) xxx–xxx 5

Table 2
LCIA midpoint assessment results by using the ReCiPe method.

Category Unit ET-D ET-ND

Amount Process Amount Process
Climate change kg CO2 eq 125.15 Electricity (54.8%) + landfill (37.3%) 68.53 Electricity (100%)
Ozone depletion kg CFC-11 eq 8.29  107 Electricity (10.4%) + wastewater (80.3%) 8.65  108 Electricity (100%)
Human toxicity kg 1,4-DB eq 4.15 Electricity (80.3%) + wastewater (13.7%) 84.37 Direct emission (96.05%)
Photochemical oxidant formation kg NMVOC 0.54 Electricity (89.8%) 0.58 Electricity (83.2%)
Particulate matter formation kg PM10 eq 0.14 Electricity (79.2%) + landfill (12.4%) 0.13 Electricity (83.0%)
Ionising radiation kg U235 eq 1.76 Wastewater (69.1%) + electricity (14.5%) + landfill (15.3%) 0.26 Electricity (100%)
Terrestrial acidification kg SO2 eq 0.35 Electricity (81.2%) 0.34 Electricity (82.7%)
Freshwater eutrophication kg P eq 5.49  103 Wastewater (43.4%) + electricity (46.8%) 2.57  103 Electricity (100%)
Marine eutrophication kg N eq 7.68  102 Electricity (77.6%) + landfill (12.4%) 7.22  102 Electricity (82.5%)
Terrestrial ecotoxicity kg 1,4-DB eq 1.08  103 Wastewater (49.1%) + electricity (35.0%) 37.83 Direct emission (100%)
Freshwater ecotoxicity kg 1,4-DB eq 8.67  102 Electricity (82.1%) 4.39 Direct emission (98.4%)
Marine ecotoxicity kg 1,4-DB eq 8.90  102 Electricity (77.8%) + wastewater (18.0%) 2.25 Direct emission (96.9%)
Agricultural land occupation m2 a 0.46 Landfill (58.7%) + Wastewater (35.2%) 8.74  103 Electricity (100%)
Urban land occupation m2 a 1.04 Electric waste (77.3%) 0.89 Electric waste (89.6%)
Natural land transformation m2 1.70  103 Electricity (13.4%) + landfill (34.1%) + wastewater (51.5%) 2.27  104 Electricity (100%)
Water depletion m3 0.16 Wastewater (70%) + landfill (24.9%) 7.17  103 Electricity (100%)
Metal depletion kg Fe eq 0.56 Wastewater (68.3%) + electricity (17.9%) + landfill (10.7%) 0.10 Electricity (100%)
Fossil depletion kg oil eq 21.93 Electricity (79.2%) + landfill (13.9%) 17.35 Electricity (100%)

0.08
(a)
Normalization

0.06
(a) Others Crude oil
0.04 180% Natural gas Coal
Bromine Beryllium
0.02 Nickel Vanadium
160% Ammonium Nitrate
0 Phosphate Particulates
140%
Substance contribution
Sulfur dioxide NMVOC
Nitrogen oxides Selenium
120% Mercury Arsenic
Methane Carbon dioxide
100%

80%

60%
7 40%
6 (b)
Normalization

5 20%
4
3 0%
CC
HT
POF
PMF
TA
FEU
MEU
FEC
MEC
FD
2
1
0
(b)
Others Nickel
Copper Mercury
120%
Zinc Silver
Antimony Barium
100%
Substance contribution

Fig. 2. Normalized LCIA mid-point results (a) ET-D; (b) ET-ND. 80%

60%
3.4. Mass balance

The mass balance of both scenarios was studied to further 40%

understand the reliability of the life cycle inventory (Fig. 4). For
both scenarios, the initial e-waste treatment mass was 1 t. The 20%
masses of products (e.g., metal, plastic, glass) from disassembly,
CRT treatment, crush, and electrodialysis plus the precious metal
0%
refining processes were 0.75 t, 9.38  103 t, 3.79  102 t, and HT TEC FEC MEC
1.66  104 t, respectively. For the ET-D scenario, the masses of
direct air emission, water emission, and solid waste disposal were Fig. 3. Contribution of substances to the key categories (a) ET-D; (b) ET-ND. CC:
climate change; HT: human toxicity; POF: photochemical oxidant formation; PMF:
7.57  106 t, 1.88  106 t, and 0.184 t, respectively. For the ET-
particulate matter formation; TA: terrestrial acidification; FEU: freshwater eutro-
ND scenario, these were 5.30  102 t, 2.80  105 t, and 0.186 t, phication; MEU: marine eutrophication; FEC: freshwater ecotoxicity; MEC: marine
respectively. Accordingly, approximately 1.72  102 t of loss for ecotoxicity; TEC: terrestrial ecotoxicity; FD: fossil depletion.

Please cite this article in press as: Hong, J., et al. Life cycle assessment of electronic waste treatment. Waste Management (2015), http://dx.doi.org/10.1016/
j.wasman.2014.12.022
6 J. Hong et al. / Waste Management xxx (2015) xxx–xxx

Operation stage Waste disposal stage

Ferrum (0.22 t); Copper Direct pollutant
(0.10 t); Aluminum (0.19 t); emissions to air:
Disassembly Plastic (0.20 t); Glass (0.04 7.57 h10-6 t
t) Direct pollutant
ET-D emissions to water:
Ferrum (9.3h10-4 t); 1.88 h10-6 t
Copper (2.52h10-3 t); Flat Solid waste to MSW
glass (4.2h10-3 t); Cone- landfill: 1.85 h10-1 t
CRT Treatment shaped glass (1.59h10-3 t); Solid waste to sanitary
Fluorescent powder
Classification

landfill: 6.9h10-4 t
(1.4h10-4 t) Solid waste to
Electric waste
(1 t) incineration: 3.31h10-3 t
Plastic (9.8h10-4 t);
Aluminum (3.39 h10-3 t);
Copper (2.77 h10-3 t);
Crush/Sorting Direct pollutant emissions
Ferrum (7.39h10-3 t); Zinc
ET-ND to air:
(9.24h10-3 t); Glass
5.30 h10-2 t
(1.41h10-2 t)
Direct pollutant emissions
to water:
Electrodialysis Platinum(5.1 h10-7t); Gold 2.80 h10-5 t
&Precious (2.04h10-5 t); Silver Solid waste open dumping:
metal refining (1.3h10-4 t); Palladium 1.32 h10-1 t
(1.53 h10-5t)

Fig. 4. Mass balance.

the ET-D scenario and 1.76  102 t of loss for the ET-ND scenario ET-ND scenario, wastewater disposal process has the lowest
can be observed, which may have resulted from missing inventory variability.
data and measurement error problems.
3.6. Uncertainty analysis

3.5. Sensitivity analysis
Table 3 shows the sensitivity analysis results obtained from the
study. A 5% decrease in electricity consumption obtains approxi-
mately 3.43 kg CO2 eq, 0.17 kg 1,4-DB eq, and 2.41  102 kg
NMVOC environmental benefit in the climate change, human tox-
icity, and photochemical oxidant formation categories, respec-
tively. For the rest categories and processes, a similar analogy
can be made with the sensitivity results shown in Table 3. Electric-
ity consumption efficiency has the highest environmental benefit
in all dominant categories, except human toxicity, terrestrial eco-
toxicity, freshwater ecotoxicity, and marine ecotoxicity. Direct soil
emissions in the ET-ND scenario produce the highest variability. By
contrast, for the ET-D scenario, solid waste disposal to incineration
has the lowest variability in all dominant categories, except cli-
mate change and marine eutrophication, in which wastewater
treatment has the lowest variability in this scenario. For the
Uncertainty analysis was conducted to estimate the degree of
confidence when the effect of the ET-D scenario is predicted to
be lower than that of the ET-ND scenario. The GSD2 and probability
results obtain with the Monte Carlo method are shown in Table 4.
For LCIA, we focus on the key categories identified in Fig. 2. The
Monte Carlo method yields a GSD2 on the climate change score
of 1.2 for ET-D and 1.4 for ET-ND. These findings indicate that
the 95% upper and lower confidence limits are the median (shown
in Table 2) multiplied and divided by GSD2, respectively. The prob-
ability that the ET-D scenario has a higher climate change score
than ET-ND is 100%. This result implies that the climate change
score of the ET-D scenario is significantly higher than that of ET-
ND scenario. For the other categories, a similar analogy can be
applied with the probability results (Table 4). In summary, the cli-
mate change, freshwater eutrophication, and fossil depletion
scores obtained from the ET-D scenario are significantly higher
than those obtained from the ET-ND. By contrast, the human

Table 3
Sensitivity of main contributors. Values are presented per functional unit.

Electricity Wastewater Solid waste

ET-D ET-ND ET-D ET-ND
Landfill Incineration
Variation 5% 5% 5% 5% 5% 5%
Climate change (kg CO2 eq) 3.43 0.23 0 2.34 0.26 0
Human toxicity (kg 1,4-DB eq) 0.17 2.89  102 3.52  104 4.76  103 7.19  103 4.05
Photochemical oxidant formation (kg NMVOC) 2.41  102 6.39  104 0 1.76  103 1.74  104 0
Particulate matter formation (kg PM10 eq) 5.37  103 4.85  104 0 8.42  104 4.49  105 0
Terrestrial acidification (kg SO2 eq) 1.41  102 1.66  103 0 1.39  103 1.2  104 0
Freshwater eutrophication (kg P eq) 1.28  104 1.19  104 0 1.43  105 1.25  105 0
Marine eutrophication (kg N eq) 2.98  103 9.35  105 3.2  108 4.78  104 2.67  104 0
Terrestrial ecotoxicity (kg 1,4-DB eq) 1.9  105 2.66  105 0 4.51  106 4.08  106 1.89
Freshwater ecotoxicity (kg 1,4-DB eq) 3.56  103 6.44  104 9.0  107 8.05  105 4.97  105 0.22
Marine ecotoxicity (kg 1,4-DB eq) 3.46  103 8.0  104 5.0  107 1.2  104 6.84  105 0.11
Fossil depletion (kg oil eq) 0.87 6.16  102 0 0.15 1.40  102 0

Please cite this article in press as: Hong, J., et al. Life cycle assessment of electronic waste treatment. Waste Management (2015), http://dx.doi.org/10.1016/
j.wasman.2014.12.022
 J. Hong et al. / Waste Management xxx (2015) xxx–xxx
Table 4
Squared geometric standard deviation (GSD2) and probability of main categories using Monte-Carlo technology.
Category GSD2
EWT with waste disposal
Climate change 1.2
Human toxicity 2.0
Photochemical oxidant formation 1.5
Particulate matter formation 1.4
Terrestrial acidification 1.4
Freshwater eutrophication 2.6
Marine eutrophication 1.5
Terrestrial ecotoxicity 1.5
Freshwater ecotoxicity 2.7
Marine ecotoxicity 2.6
Fossil depletion 1.3
D: disposal; ND: without disposal.
toxicity, terrestrial ecotoxicity, freshwater ecotoxicity, and marine
ecotoxicity scores obtained from the ET-D scenario are significantly
lower than those obtained from ET-ND. Similar LCIA scores are
observed in both scenarios for photochemical oxidant formation,
particulate matter formation, terrestrial acidification, and marine
eutrophication.
4. Discussion
Currently, governments are increasingly turning their attention
to e-waste disposal because the increasing amount of e-waste
worldwide is a critical environmental problem. Although the LCA
of e-waste has been extensively studied (Song et al., 2012; Niu
et al., 2012), the potential environmental impact of e-waste treat-
ment widely varies (Kiddee et al., 2013). The key process that con-
tributes to the overall environmental burden for both scenarios is
electricity generation (Table 2). In this study, the electricity con-
sumption for e-waste recycling is 145.45 kW h/t-metal, which is
higher than that in the European database (125.69 kW h/t-metal,
process ID: 14601303340, Ecoinvent centre, 2010). If Europe-avail-
able technology is used in the present research, approximately
9.48% and 8.04% of the overall potential impact of electricity gen-
eration will be reduced in the ET-D and ET-ND scenarios, respec-
tively. Accordingly, improving electricity consumption efficiency
is the key to reducing the overall environmental burden for both
scenarios.
Notably, data on chemical production in Europe (Ecoinvent
centre, 2010) were used in this research because of the lack of
information on China. For both scenarios, the uncertainties of over-
all environmental burden were mainly generated from the emis-
sions of heavy metal and the consumption of electricity. Table 2
shows that the overall environmental impact is mainly generated
from coal-based electricity generation, landfill, and wastewater
disposal processes for the ET-D scenario. That of the ET-ND sce-
nario is electricity generation and direct pollutant emissions from
e-waste on-site disposal. Therefore, although European data on
chemical production were used in this study, the overall environ-
mental impact from both e-waste recycling scenarios was not
affected.
Table 4 shows that the overall environmental burden obtained
from the ET-ND scenario is significantly higher than that obtained
from the ET-D because of the relatively high environmental impact
of human toxicity, terrestrial ecotoxicity, freshwater ecotoxicity,
and marine ecotoxicity obtained from the former scenario (Table 4).
Tsydenova and Bengtsson (2011) reported that the improper han-
dling and management of e-waste may pose significant human and
environmental health risks because hazards arise from heavy met-
als and halogenated compounds in e-waste. Previous studies
Please cite this article in press as: Hong, J., et al. Life cycle assessment of electronic waste treatment. Waste Management (2015), http://dx.doi.org/10.1016/
j.wasman.2014.12.022
7
Probability of EWT-D > EWT-ND
EWT without waste disposal
1.4 100%
2.4 0%
1.5 33.3%
1.5 65.4%
1.5 58.1%
7.4 99.8%
1.5 67.5%
3.0 0%
2.8 0%
2.7 0%
1.3 90.2%
reported that large amounts of e-waste in China are improperly
collected and disposed (Chi et al., 2014; Xu et al., 2014), so these
from many high-level toxic organic compounds in various environ-
mental mediums and biological samples around e-waste disposal
sites (Chi et al., 2014). Accordingly, the proper handling and man-
agement of e-waste are the key in decreasing its risk to both
human health and environment. Compared with industrialized
nations, developing countries lack the conventions, directives,
and laws that govern producer responsibility (Sthiannopkao and
Wong, 2013), which can efficiently reduce the amount of e-waste
that is improper disposal. For instance, although an e-waste dis-
posal law was enacted in China in 2009 (State council, 2009), the
obligations and penalties in the law are loosely implemented
(Sthiannopkao and Wong, 2013). Clearly defining the duties of
manufacturers, retailers, recycling companies, and consumers is
important to reducing the amount of improper e-waste disposal.
In addition, although some e-waste resources (e.g., plastic, met-
als, and glass) have significant environmental benefits compared
with their primary manufacturing processes (Song et al., 2012),
considerable environmental burden is generated during the e-
waste recycling stage (Table 2). Kiddee et al. (2013) reported that
importing e-waste and electronic goods from developed countries
causes a major e-waste problem in developing countries. Wang
et al. (2013) reported that China is now facing serious e-waste
problems as a result of foreign imports. Approximately 80% and
18% of e-waste worldwide are produced by developed countries
(e.g., Europe, U.S.A.) and China, respectively, accounting for 14%
of the national municipal solid waste (MSW) generation in China
in 2009 (China Statistics Yearbook, 2010). Therefore, if all e-waste
is recycled, the annual e-waste importation and generation in
China will cause a significant environmental burden (Table 5).
Zoeteman et al. (2010) and Ongondo et al. (2011) reported that
the use of landfills is also a commonly used e-waste disposal
method worldwide. Approximately 40% of e-waste comes from dis-
posal to landfill in China (Zoeteman et al., 2010; Hong et al.,
2010b), whereas approximately 70% of heavy metals in US landfills
come from e-waste (Widmer et al., 2005). An LCA analysis of MSW
disposal to landfill and incineration in China was reported by Hong
et al. (2010b). Fig. 5 compares the LCIA results of e-waste disposal
to landfill and incineration (without energy recovery), and the ET-
D and ET-ND scenarios. E-waste with end-life disposal to incinera-
tion has the highest environmental impact among the categories,
except for climate change, ozone depletion, ecotoxicity (i.e., mar-
ine, terrestrial, and freshwater), urban land occupation, natural
and agricultural land transformation, and water depletion
(Fig. 5a). In ozone depletion and urban land occupation, the highest
environmental burden is attributed to the ET-D scenario, whereas
that in ecotoxicity (i.e., marine, terrestrial, and freshwater) is
8 J. Hong et al. / Waste Management xxx (2015) xxx–xxx

Table 5
Annual estimated environmental burden generated by e-waste import and generation in China.

Average annual LCIA value from e-waste import Average annual LCIA value from e-waste generation
ET-D ET-ND ET-D ET-ND
Climate change (kg CO2 eq) 1.26  109 6.91  108 5.04  108 2.76  108
Human toxicity (kg 1,4-DB eq) 4.18  107 8.5  108 1.67  107 3.4  108
Photochemical oxidant formation (kg NMVOC) 5.42  106 5.84  106 2.16  106 2.33  106
Particulate matter formation (kg PM10 eq) 1.37  106 1.3  106 5.46  105 5.21  105
Terrestrial acidification (kg SO2 eq) 3.51  106 3.45  106 1.4  106 1.38  106
Freshwater eutrophication (kg P eq) 5.53  104 2.59  104 2.21  104 1.03  104
Marine eutrophication (kg N eq) 7.75  105 7.28  105 3.1  105 2.91  105
Terrestrial ecotoxicity (kg 1,4-DB eq) 1.09  104 3.81  108 4.36  103 1.52  108
Freshwater ecotoxicity (kg 1,4-DB eq) 8.98  105 4.43  107 3.49  105 1.77  107
Marine ecotoxicity (kg 1,4-DB eq) 8.98  105 2.27  107 3.59  105 9.06  106
Fossil depletion (kg oil eq) 2.21  108 1.75  108 8.83  107 6.99  107

Landfill ET-ND ET-D Incineartion

technologies, which are methods of waste disposal by burning at
100 800–900 °C and 1300–1800 °C, respectively. Their results showed
80 that waste melting technology can significantly reduce toxic pollu-
%

60 tants because of their crystallizability at high temperature. There-

40 fore, e-waste melting technology might be a better choice than e-
20 waste incineration from an environmental perspective. The overall
0 environmental impact of the ET-ND and incineration scenarios is
ge n ty n n n n n n ty ty ity n n n n on n
an tio ici tio tio tio tio io tio ici ici ic tio io io tio ti tio
ch le ox a a dia ica at ica ox x ox pa at at ple ple le mainly attributed to human toxicity, freshwater ecotoxicity, and
ate e dep an t form formg ra idif phicoph ecot ecotoecot ccu ccup formr de al de l dep
m n m t r n ac ro tr l e o o s te t si marine ecotoxicity (Fig. 5b). The impact of freshwater eutrophica-
cli ozo hu idan atteonisi rial r eut e eu stria ater arin land land tran wa me fos
x m i est te rin re hw m al n nd
al o late r r a a er es ur ba la tion and terrestrial ecotoxicity also plays an important role in the
c
mi icu te eshw m t fr ult ur tural
che part fr gric na incineration and ET-ND scenarios, respectively. LCA was conducted
a
oto
ph with the gate-to-gate approach in the current study. The environ-
mental benefit of recovering metals, glass, and plastic from e-waste
6 was excluded in this study. Song et al. (2012) reported that the
Nomalized value

Landfill
ET-ND recovery of copper and plastic from e-waste can generate
4 ET-D significant environmental benefits. Therefore, the lowest overall
Incineartion
environmental impact could be found in the ET-D scenario in con-
2 sideration of the environmental benefit of the recovery of metals,
glass, and plastic from e-waste.
0
ge n ty n n n n n n ty ty ity n n n n on n
an tio ici tio tio tio tio io tio ici ici ic tio io io tio ti tio
ch le ox a a dia ica at ica ox x ox pa at at ple ple le
ate e depan t formformg ra idif phicoph ecot ecotoecot ccu ccupformr deal de l dep 5. Conclusion
m n m t r n ac ro tr l e o o s te t si
cli ozo hu idan atteonisi rial r eut e eu stria ater arin land land tran wa me fos
x m i est te rin re hw m al n nd
l o late r a a r s r a a
c a r
te eshw m te fre lt rb l l
u
mi icu cu uatura This study compared the LCIA of e-waste with and without end-
c he part fr agri n
t o
ph
o life disposal. To add credibility to the study, sensitivity and uncer-
tainty analyses were also conducted. The life cycle inventory, the
Fig. 5. Life cycle impact assessment results comparison. (a) Midpoint results and key factors identified, and the LCIA analysis results will be helpful
(b) normalized value.
to e-waste management decision makers. The main findings
showed that although the impact of the ET-D scenario for climate
change, freshwater eutrophication, and fossil depletion are signifi-
attributed to the ET-ND scenario. For the other categories, landfill
cantly higher than those obtained from the ET-ND scenario, the
technology was the highest contributor. The normalized overall
overall environmental impact of the ET-ND scenario is significantly
environmental impact of landfill, incineration, ET-D, and ET-ND
higher than that of the ET-D scenario. This disadvantage compared
scenarios is 0.10, 5.41, 0.24, and 9.38, respectively. The overall
with that of the ET-D scenario is mainly ascribed to the increase in
environmental impact of the incineration scenario is approxi-
direct soil emissions during the waste open dumping stage. There-
mately 54 times higher than that of landfill because of the direct
fore, the scientific improvement of the end-life disposal process of
air emission of heavy metals. Song et al. (2012) compared e-waste
e-waste treatment and the reduction of improper e-waste treat-
disposal to landfill and incineration technologies by using
ment amounts are efficient ways to reduce the overall environ-
‘‘Eco-indicator 99’’ method on the basis of the Ecoinvent database.
mental burden. Compared with the ET-D scenario, incineration
The results showed that the overall environmental impact from the
and landfill technologies are also not preferred for e-waste disposal
incineration scenario is approximately 37 times higher than that
because e-waste can produce many important mineral resources.
from landfill. The difference in overall environmental burden
However, the environmental benefits from the recovery of those
between incineration and landfill obtained from this study is larger
mineral resources are unclear. Further research on this subject is
than that reported by Song et al. (2012) because of regionalization,
therefore needed.
uncertainties, and geographical variability in the applied invento-
ries and LCIA models. The energy type and system boundaries con-
sidered in each research played additional and important roles in Acknowledgments
the variation. Niu et al. (2012) also proved that e-waste
incineration can generate a significant environmental burden com- We gratefully acknowledge financial support from supported
pared with MSW incineration because of direct toxic material by Program for New Century Excellent Talents in University of
emissions. Hong et al. (2009) compared incineration and melting China (NCET-13-0344), the institute of plateau meteorology,

Please cite this article in press as: Hong, J., et al. Life cycle assessment of electronic waste treatment. Waste Management (2015), http://dx.doi.org/10.1016/
j.wasman.2014.12.022
 J. Hong et al. / Waste Management xxx (2015) xxx–xxx
CMA, Chengdu, China (LPM2014002), the National Natural Science
Foundation of China (Grant No. 41101554), and national high-tech
R&D program of China (863 program, Grant No. 2012AA061705).
References
Allsopp, M., Santillo, D., Johnston, P., 2006. Environmental and human health
concerns in the processing of electrical and electronic waste. Greenpeace
research laboratories, department of biological sciences, University of Exeter
EX4 4PS, UK.
Brigden, K., Labunska, I., Santillo, D., Allsopp, M., 2005. Recycling of electronic
wastes in China and India: Workplace and Environmental Contamination.
Greenpeace International, Technical Note: 09/2005 (Section 1).
Chen, W., Hong, J., Xu, C., 2014. Pollutants generated by cement production in China,
their impacts, and the potential for environmental improvement. J. Clean. Prod.
00, <http://www.sciencedirect.com/science/article/pii/S095965261400403X>.
Chi, X., Wang, M., Reuter, M., 2014. E-waste collection channels and household
recycling behaviors in Taizhou of China. J. Clean. Prod. 80, 87–95.
China Statistics Yearbook, 2010. National bureau of statistic of China. <http://www.
stats.gov.cn/tjsj/ndsj/2010/indexch.htm>.
Cui, X., Hong, J., Gao, M., 2012. Environmental impact assessment of three coal-
based electricity generation scenarios in China. Energy 45, 952–959.
Ecoinvent centre, 2010. Ecoinvent database v2.2. Swiss center for life cycle
inventories. <http://www.ecoinvent.org/database/>.
Goedkoop, M., Heijungs, R., Huijbregts, M., De Schryver, A., Struijs, J., Van Zelm, R.,
2009. ReCiPe 2008-A Life Cycle Impact Assessment Method which Comprises
Harmonised Category Indicators at the Midpoint and the Endpoint Level. First
edition Report I: Characterization. DenHaag, The Netherlands.
Hong, J., Hong, J., Otaki, M., Olivier, J., 2009. Environmental and economic life cycle
assessment for sewage sludge treatment processes in Japan. Waste Manage. 29,
696–703.
Hong, J., Shaked, S., Rosenbaum, R., Jolliet, O., 2010a. Analytical uncertainty
propagation in life cycle inventory and impact assessment: application to an
automobile front panel. Int. J. Life Cycle Assess. 15, 499–510.
Hong, J., Li, X., Cui, Z., 2010b. Life cycle assessment of four municipal solid waste
management scenarios in China. Waste Manage. 30, 2362–2369.
ISO (International Organization for Standardization), 2006. Environmental
Management-Life Cycle Assessment-General Principles and Framework. ISO
14040, Geneva, Switzerland.
Kiddee, P., Naidu, R., Hong, M., 2013. Electronic waste management approaches: an
overview. Waste Manage. 33, 1237–1250.
Please cite this article in press as: Hong, J., et al. Life cycle assessment of electronic waste treatment. Waste Management (2015), http://dx.doi.org/10.1016/
j.wasman.2014.12.022
9
Li, Y., Duan, Y., Huang, F., Yang, J., Xiang, N., Meng, X., et al., 2014. Polybrominated
diphenyl ethers in e-waste: level and transfer in a typical e-waste recycling site
in Shanghai, Eastern China. Waste Manage. 34, 1059–1065.
Lin, W., Liu, Y., 2012. Present status of e-waste disposal and recycling in China. Proc.
Environ. Sci. 16, 506–514.
Liu, Q., Cao, J., Li, K., Miao, X., Li, G., Fan, F., et al., 2009. Chromosomal aberrations
and DNA damage in human populations exposed to the processing of
electronics waste. Environ. Sci. Pollut. Res. 16, 329–338.
Niu, R., Wang, Z., Song, Q., Li, J., 2012. LCA of scrap CRT display at various scenarios
of treatment. Proc. Environ. Sci. 16, 576–584.
Ongondo, F.O., Williams, I.D., Cherrett, T.J., 2011. How are WEEE doing? A global
review of the management of electrical and electronic wastes. Waste Manage.
31, 714–730.
Qu, Y., Zhu, Q., Joseph, S., Geng, Y., Zhong, Y., 2013. A review of developing an e-
wastes collection system in Dalian, China. J. Clean. Prod. 52, 176–184.
Robinson, B., 2009. E-waste: an assessment of global production and environmental
impacts. Sci. Total Environ. 408, 183–191.
Song, Q., Wang, Z., Li, J., Zeng, X., 2012. Life cycle assessment of TV sets in China: a
case study of the impacts of CRT monitors. Waste Manage. 32, 1926–1936.
State Council, 2009. Regulations for the administration of the recovery and disposal
of electric and electronic products. Order of the State Council of the People’s
Republic of China, No. 551. <http://www.lawinfochina.com/display.aspx? id=
7368&lib=law>.
Sthiannopkao, S., Wong, M., 2013. Handling e-waste in developed and developing
countries: Initiatives, practices, and consequences. Sci. Total Environ. 463–464,
1147–1153.
Tsydenova, O., Bengtsson, M., 2011. Chemical hazards associated with treatment of
waste electrical and electronic equipment. Waste Manage. 31, 45–58.
Wang, F., Kuehr, R., Ahiquist, D., Li, J., 2013. E-waste in China: a country report. StEP
Green Paper Series. ISSN:2219-6560. <http://www.step-initiative.org/tl_files/step/
files/5.%20StEP%20Task%20Forces/5.1%20TF%201%20Policy/5.1.4%20Projects/
StEP%20Country%20Study%20in%20China/StEP%20GP_E-waste%20China.pdf>.
Wegener Sleeswijk, A., Van Oers, L., Guinée, J., Struijs, J., Huijbregts, M., 2008.
Normalisation in product life cycle assessment: An LCA of the global and
European economic systems in the year 2000. Sci. Total Environ. 390, 227–240.
Widmer, R., Oswald-Krapf, H., Sinha-Khetriwal, D., Schnellmann, M., Boni, H., 2005.
Global perspectives on e-waste. Environ. Impact Assess. 25, 436–458.
Xu, P., Lou, X., Ding, G., Shen, H., Wu, L., Chen, Z., et al., 2014. Association of PCB,
PBDE and PCDD/F body burdens with hormone levels for children in an e-waste
dismantling area of Zhejiang Province, China. Sci. Total Environ. 499, 55–61.
Zoeteman, C., Haarold, R., Jan, V., 2010. Handling WEEE waste flows: on the
effectiveness of producer responsibility in a globalizing world. Int. J. Adv.
Manuf. Technol. 47, 415–436.